Graph Federated Learning(GFL)has shown great potential in privacy protection and distributed intelligence through distributed collaborative training of graph-structured data without sharing raw information.However,exi...Graph Federated Learning(GFL)has shown great potential in privacy protection and distributed intelligence through distributed collaborative training of graph-structured data without sharing raw information.However,existing GFL approaches often lack the capability for comprehensive feature extraction and adaptive optimization,particularly in non-independent and identically distributed(NON-IID)scenarios where balancing global structural understanding and local node-level detail remains a challenge.To this end,this paper proposes a novel framework called GFL-SAR(Graph Federated Collaborative Learning Framework Based on Structural Amplification and Attention Refinement),which enhances the representation learning capability of graph data through a dual-branch collaborative design.Specifically,we propose the Structural Insight Amplifier(SIA),which utilizes an improved Graph Convolutional Network(GCN)to strengthen structural awareness and improve modeling of topological patterns.In parallel,we propose the Attentive Relational Refiner(ARR),which employs an enhanced Graph Attention Network(GAT)to perform fine-grained modeling of node relationships and neighborhood features,thereby improving the expressiveness of local interactions and preserving critical contextual information.GFL-SAR effectively integrates multi-scale features from every branch via feature fusion and federated optimization,thereby addressing existing GFL limitations in structural modeling and feature representation.Experiments on standard benchmark datasets including Cora,Citeseer,Polblogs,and Cora_ML demonstrate that GFL-SAR achieves superior performance in classification accuracy,convergence speed,and robustness compared to existing methods,confirming its effectiveness and generalizability in GFL tasks.展开更多
Automated classification of gas flow states in blast furnaces using top-camera imagery typically demands a large volume of labeled data,whose manual annotation is both labor-intensive and cost-prohibitive.To mitigate ...Automated classification of gas flow states in blast furnaces using top-camera imagery typically demands a large volume of labeled data,whose manual annotation is both labor-intensive and cost-prohibitive.To mitigate this challenge,we present an enhanced semi-supervised learning approach based on the Mean Teacher framework,incorporating a novel feature loss module to maximize classification performance with limited labeled samples.The model studies show that the proposed model surpasses both the baseline Mean Teacher model and fully supervised method in accuracy.Specifically,for datasets with 20%,30%,and 40%label ratios,using a single training iteration,the model yields accuracies of 78.61%,82.21%,and 85.2%,respectively,while multiple-cycle training iterations achieves 82.09%,81.97%,and 81.59%,respectively.Furthermore,scenario-specific training schemes are introduced to support diverse deployment need.These findings highlight the potential of the proposed technique in minimizing labeling requirements and advancing intelligent blast furnace diagnostics.展开更多
Accurate prediction of flood events is important for flood control and risk management.Machine learning techniques contributed greatly to advances in flood predictions,and existing studies mainly focused on predicting...Accurate prediction of flood events is important for flood control and risk management.Machine learning techniques contributed greatly to advances in flood predictions,and existing studies mainly focused on predicting flood resource variables using single or hybrid machine learning techniques.However,class-based flood predictions have rarely been investigated,which can aid in quickly diagnosing comprehensive flood characteristics and proposing targeted management strategies.This study proposed a prediction approach of flood regime metrics and event classes coupling machine learning algorithms with clustering-deduced membership degrees.Five algorithms were adopted for this exploration.Results showed that the class membership degrees accurately determined event classes with class hit rates up to 100%,compared with the four classes clustered from nine regime metrics.The nonlinear algorithms(Multiple Linear Regression,Random Forest,and least squares-Support Vector Machine)outperformed the linear techniques(Multiple Linear Regression and Stepwise Regression)in predicting flood regime metrics.The proposed approach well predicted flood event classes with average class hit rates of 66.0%-85.4%and 47.2%-76.0%in calibration and validation periods,respectively,particularly for the slow and late flood events.The predictive capability of the proposed prediction approach for flood regime metrics and classes was considerably stronger than that of hydrological modeling approach.展开更多
Vehicle Edge Computing(VEC)and Cloud Computing(CC)significantly enhance the processing efficiency of delay-sensitive and computation-intensive applications by offloading compute-intensive tasks from resource-constrain...Vehicle Edge Computing(VEC)and Cloud Computing(CC)significantly enhance the processing efficiency of delay-sensitive and computation-intensive applications by offloading compute-intensive tasks from resource-constrained onboard devices to nearby Roadside Unit(RSU),thereby achieving lower delay and energy consumption.However,due to the limited storage capacity and energy budget of RSUs,it is challenging to meet the demands of the highly dynamic Internet of Vehicles(IoV)environment.Therefore,determining reasonable service caching and computation offloading strategies is crucial.To address this,this paper proposes a joint service caching scheme for cloud-edge collaborative IoV computation offloading.By modeling the dynamic optimization problem using Markov Decision Processes(MDP),the scheme jointly optimizes task delay,energy consumption,load balancing,and privacy entropy to achieve better quality of service.Additionally,a dynamic adaptive multi-objective deep reinforcement learning algorithm is proposed.Each Double Deep Q-Network(DDQN)agent obtains rewards for different objectives based on distinct reward functions and dynamically updates the objective weights by learning the value changes between objectives using Radial Basis Function Networks(RBFN),thereby efficiently approximating the Pareto-optimal decisions for multiple objectives.Extensive experiments demonstrate that the proposed algorithm can better coordinate the three-tier computing resources of cloud,edge,and vehicles.Compared to existing algorithms,the proposed method reduces task delay and energy consumption by 10.64%and 5.1%,respectively.展开更多
In the realm of Intelligent Railway Transportation Systems,effective multi-party collaboration is crucial due to concerns over privacy and data silos.Vertical Federated Learning(VFL)has emerged as a promising approach...In the realm of Intelligent Railway Transportation Systems,effective multi-party collaboration is crucial due to concerns over privacy and data silos.Vertical Federated Learning(VFL)has emerged as a promising approach to facilitate such collaboration,allowing diverse entities to collectively enhance machine learning models without the need to share sensitive training data.However,existing works have highlighted VFL’s susceptibility to privacy inference attacks,where an honest but curious server could potentially reconstruct a client’s raw data from embeddings uploaded by the client.This vulnerability poses a significant threat to VFL-based intelligent railway transportation systems.In this paper,we introduce SensFL,a novel privacy-enhancing method to against privacy inference attacks in VFL.Specifically,SensFL integrates regularization of the sensitivity of embeddings to the original data into the model training process,effectively limiting the information contained in shared embeddings.By reducing the sensitivity of embeddings to the original data,SensFL can effectively resist reverse privacy attacks and prevent the reconstruction of the original data from the embeddings.Extensive experiments were conducted on four distinct datasets and three different models to demonstrate the efficacy of SensFL.Experiment results show that SensFL can effectively mitigate privacy inference attacks while maintaining the accuracy of the primary learning task.These results underscore SensFL’s potential to advance privacy protection technologies within VFL-based intelligent railway systems,addressing critical security concerns in collaborative learning environments.展开更多
The presence of aluminum(Al^(3+))and fluoride(F^(−))ions in the environment can be harmful to ecosystems and human health,highlighting the need for accurate and efficient monitoring.In this paper,an innovative approac...The presence of aluminum(Al^(3+))and fluoride(F^(−))ions in the environment can be harmful to ecosystems and human health,highlighting the need for accurate and efficient monitoring.In this paper,an innovative approach is presented that leverages the power of machine learning to enhance the accuracy and efficiency of fluorescence-based detection for sequential quantitative analysis of aluminum(Al^(3+))and fluoride(F^(−))ions in aqueous solutions.The proposed method involves the synthesis of sulfur-functionalized carbon dots(C-dots)as fluorescence probes,with fluorescence enhancement upon interaction with Al^(3+)ions,achieving a detection limit of 4.2 nmol/L.Subsequently,in the presence of F^(−)ions,fluorescence is quenched,with a detection limit of 47.6 nmol/L.The fingerprints of fluorescence images are extracted using a cross-platform computer vision library in Python,followed by data preprocessing.Subsequently,the fingerprint data is subjected to cluster analysis using the K-means model from machine learning,and the average Silhouette Coefficient indicates excellent model performance.Finally,a regression analysis based on the principal component analysis method is employed to achieve more precise quantitative analysis of aluminum and fluoride ions.The results demonstrate that the developed model excels in terms of accuracy and sensitivity.This groundbreaking model not only showcases exceptional performance but also addresses the urgent need for effective environmental monitoring and risk assessment,making it a valuable tool for safeguarding our ecosystems and public health.展开更多
Liposomes serve as critical carriers for drugs and vaccines,with their biological effects influenced by their size.The microfluidic method,renowned for its precise control,reproducibility,and scalability,has been wide...Liposomes serve as critical carriers for drugs and vaccines,with their biological effects influenced by their size.The microfluidic method,renowned for its precise control,reproducibility,and scalability,has been widely employed for liposome preparation.Although some studies have explored factors affecting liposomal size in microfluidic processes,most focus on small-sized liposomes,predominantly through experimental data analysis.However,the production of larger liposomes,which are equally significant,remains underexplored.In this work,we thoroughly investigate multiple variables influencing liposome size during microfluidic preparation and develop a machine learning(ML)model capable of accurately predicting liposomal size.Experimental validation was conducted using a staggered herringbone micromixer(SHM)chip.Our findings reveal that most investigated variables significantly influence liposomal size,often interrelating in complex ways.We evaluated the predictive performance of several widely-used ML algorithms,including ensemble methods,through cross-validation(CV)for both lipo-some size and polydispersity index(PDI).A standalone dataset was experimentally validated to assess the accuracy of the ML predictions,with results indicating that ensemble algorithms provided the most reliable predictions.Specifically,gradient boosting was selected for size prediction,while random forest was employed for PDI prediction.We successfully produced uniform large(600 nm)and small(100 nm)liposomes using the optimised experimental conditions derived from the ML models.In conclusion,this study presents a robust methodology that enables precise control over liposome size distribution,of-fering valuable insights for medicinal research applications.展开更多
As an effective strategy to address urban traffic congestion,traffic flow prediction has gained attention from Federated-Learning(FL)researchers due FL’s ability to preserving data privacy.However,existing methods fa...As an effective strategy to address urban traffic congestion,traffic flow prediction has gained attention from Federated-Learning(FL)researchers due FL’s ability to preserving data privacy.However,existing methods face challenges:some are too simplistic to capture complex traffic patterns effectively,and others are overly complex,leading to excessive communication overhead between cloud and edge devices.Moreover,the problem of single point failure limits their robustness and reliability in real-world applications.To tackle these challenges,this paper proposes a new method,CMBA-FL,a Communication-Mitigated and Blockchain-Assisted Federated Learning model.First,CMBA-FL improves the client model’s ability to capture temporal traffic patterns by employing the Encoder-Decoder framework for each edge device.Second,to reduce the communication overhead during federated learning,we introduce a verification method based on parameter update consistency,avoiding unnecessary parameter updates.Third,to mitigate the risk of a single point of failure,we integrate consensus mechanisms from blockchain technology.To validate the effectiveness of CMBA-FL,we assess its performance on two widely used traffic datasets.Our experimental results show that CMBA-FL reduces prediction error by 11.46%,significantly lowers communication overhead,and improves security.展开更多
Objective Accurately identifying the key influencing factors of psychological birth trauma in primiparous women is crucial for implementing effective preventive and intervention measures.This study aimed to develop an...Objective Accurately identifying the key influencing factors of psychological birth trauma in primiparous women is crucial for implementing effective preventive and intervention measures.This study aimed to develop and validate an interpretable machine learning prediction model for identifying the key influencing factors of psychological birth trauma in primiparous women.Methods A multicenter cross-sectional study was conducted on primiparous women in four tertiary hospitals in Sichuan Province,southwestern China,from December 2023 to March 2024.The Childbirth Trauma Index was used in assessing psychological birth trauma in primiparous women.Data were collected and randomly divided into a training set(80%,n=289)and a testing set(20%,n=73).Six different machine learning models were trained and tested.Training and prediction were conducted using six machine learning models included Linear Regression,Support Vector Regression,Multilayer Perceptron Regression,eXtreme Gradient Boosting Regression,Random Forest Regression,and Adaptive Boosting Regression.The optimal model was selected based on various performance metrics,and its predictive results were interpreted using SHapley Additive exPlanations(SHAP)and accumulated local effects(ALE).Results Among the six machine learning models,the Multilayer Perceptron Regression model exhibited the best overall performance in the testing set(MAE=3.977,MSE=24.832,R2=0.507,EVS=0.524,RMSE=4.983).In the testing set,the R2 and EVS of the Multilayer Perceptron Regression model increased by 8.3%and 1.2%,respectively,compared to the traditional linear regression model.Meanwhile,the MAE,MSE,and RMSE decreased by 0.4%,7.3%,and 3.7%,respectively,compared to the traditional linear regression model.The SHAP analysis indicated that intrapartum pain,anxiety,postpartum pain,resilience,and planned pregnancy are the most critical influencing factors of psychological birth trauma in primiparous women.The ALE analysis indicated that higher intrapartum pain,anxiety,and postpartum pain scores are risk factors,while higher resilience scores are protective factors.Conclusions Interpretable machine learning prediction models can identify the key influencing factors of psychological birth trauma in primiparous women.SHAP and ALE analyses based on the Multilayer Perceptron Regression model can help healthcare providers understand the complex decision-making logic within a prediction model.This study provides a scientific basis for the early prevention and personalized intervention of psychological birth trauma in primiparous women.展开更多
In order to study the characteristics of pure fly ash-based geopolymer concrete(PFGC)conveniently,we used a machine learning method that can quantify the perception of characteristics to predict its compressive streng...In order to study the characteristics of pure fly ash-based geopolymer concrete(PFGC)conveniently,we used a machine learning method that can quantify the perception of characteristics to predict its compressive strength.In this study,505 groups of data were collected,and a new database of compressive strength of PFGC was constructed.In order to establish an accurate prediction model of compressive strength,five different types of machine learning networks were used for comparative analysis.The five machine learning models all showed good compressive strength prediction performance on PFGC.Among them,R2,MSE,RMSE and MAE of decision tree model(DT)are 0.99,1.58,1.25,and 0.25,respectively.While R2,MSE,RMSE and MAE of random forest model(RF)are 0.97,5.17,2.27 and 1.38,respectively.The two models have high prediction accuracy and outstanding generalization ability.In order to enhance the interpretability of model decision-making,we used importance ranking to obtain the perception of machine learning model to 13 variables.These 13 variables include chemical composition of fly ash(SiO_(2)/Al_(2)O_(3),Si/Al),the ratio of alkaline liquid to the binder,curing temperature,curing durations inside oven,fly ash dosage,fine aggregate dosage,coarse aggregate dosage,extra water dosage and sodium hydroxide dosage.Curing temperature,specimen ages and curing durations inside oven have the greatest influence on the prediction results,indicating that curing conditions have more prominent influence on the compressive strength of PFGC than ordinary Portland cement concrete.The importance of curing conditions of PFGC even exceeds that of the concrete mix proportion,due to the low reactivity of pure fly ash.展开更多
Teacher–student relationships play a vital role in improving college students’academic performance and the quality of higher education.However,empirical studies with substantial data-driven insights remain limited.T...Teacher–student relationships play a vital role in improving college students’academic performance and the quality of higher education.However,empirical studies with substantial data-driven insights remain limited.To address this gap,this study collected 3278 questionnaires from seven universities across four provinces in China to analyze the key factors affecting college students’academic performance.A machine learning framework,CQFOA-KELM,was developed by enhancing the Fruit Fly Optimization Algorithm(FOA)with Covariance Matrix Adaptation Evolution Strategy(CMAES)and Quadratic Approximation(QA).CQFOA significantly improved population diversity and was validated on the IEEE CEC2017 benchmark functions.The CQFOA-KELM model achieved an accuracy of 98.15%and a sensitivity of 98.53%in predicting college students’academic performance.Additionally,it effectively identified the key factors influencing academic performance through the feature selection process.展开更多
The distributed permutation flow shop scheduling problem(DPFSP)has received increasing attention in recent years.The iterated greedy algorithm(IGA)serves as a powerful optimizer for addressing such a problem because o...The distributed permutation flow shop scheduling problem(DPFSP)has received increasing attention in recent years.The iterated greedy algorithm(IGA)serves as a powerful optimizer for addressing such a problem because of its straightforward,single-solution evolution framework.However,a potential draw-back of IGA is the lack of utilization of historical information,which could lead to an imbalance between exploration and exploitation,especially in large-scale DPFSPs.As a consequence,this paper develops an IGA with memory and learning mechanisms(MLIGA)to efficiently solve the DPFSP targeted at the mini-malmakespan.InMLIGA,we incorporate a memory mechanism to make a more informed selection of the initial solution at each stage of the search,by extending,reconstructing,and reinforcing the information from previous solutions.In addition,we design a twolayer cooperative reinforcement learning approach to intelligently determine the key parameters of IGA and the operations of the memory mechanism.Meanwhile,to ensure that the experience generated by each perturbation operator is fully learned and to reduce the prior parameters of MLIGA,a probability curve-based acceptance criterion is proposed by combining a cube root function with custom rules.At last,a discrete adaptive learning rate is employed to enhance the stability of the memory and learningmechanisms.Complete ablation experiments are utilized to verify the effectiveness of the memory mechanism,and the results show that this mechanism is capable of improving the performance of IGA to a large extent.Furthermore,through comparative experiments involving MLIGA and five state-of-the-art algorithms on 720 benchmarks,we have discovered that MLI-GA demonstrates significant potential for solving large-scale DPFSPs.This indicates that MLIGA is well-suited for real-world distributed flow shop scheduling.展开更多
Attempts have been made to modulate motor sequence learning(MSL)through repetitive transcranial magnetic stimulation,targeting different sites within the sensorimotor network.However,the target with the optimum modula...Attempts have been made to modulate motor sequence learning(MSL)through repetitive transcranial magnetic stimulation,targeting different sites within the sensorimotor network.However,the target with the optimum modulatory effect on neural plasticity associated with MSL remains unclarified.This study was therefore designed to compare the role of the left primary motor cortex and the left supplementary motor area proper(SMAp)in modulating MSL across different complexity levels and for both hands,as well as the associated neuroplasticity by applying intermittent theta burst stimulation together with the electroencephalogram and concurrent transcranial magnetic stimulation.Our data demonstrated the role of SMAp stimulation in modulating neural communication to support MSL,which is achieved by facilitating regional activation and orchestrating neural coupling across distributed brain regions,particularly in interhemispheric connections.These findings may have important clinical implications,particularly for motor rehabilitation in populations such as post-stroke patients.展开更多
While artificial intelligence(AI)shows promise in education,its real-world effectiveness in specific settings like blended English as a Foreign Language(EFL)learning needs closer examination.This study investigated th...While artificial intelligence(AI)shows promise in education,its real-world effectiveness in specific settings like blended English as a Foreign Language(EFL)learning needs closer examination.This study investigated the impact of a blended teaching model incorporating AI tools on the Superstar Learning Platform for Chinese university EFL students.Using a mixed-methods approach,60 first-year students were randomized into an experimental group(using the AI-enhanced model)and a control group(traditional instruction)for 16 weeks.Data included test scores,learning behaviors(duration,task completion),satisfaction surveys,and interviews.Results showed the experimental group significantly outperformed the control group on post-tests and achieved larger learning gains.These students also demonstrated greater engagement through longer study times and higher task completion rates,and reported significantly higher satisfaction.Interviews confirmed these findings,with students attributing benefits to the model’s personalized guidance,structured content presentation(knowledge graphs),immediate responses,flexibility,and varied interaction methods.However,limitations were noted,including areas where the platform’s AI could be improved(e.g.,for assessing speaking/translation)and ongoing challenges with student self-discipline.The study concludes that this AI-enhanced blended model significantly improved student performance,engagement,and satisfaction in this EFL context.The findings offer practical insights for educators and platform developers,suggesting AI integration holds significant potential while highlighting areas for refinement.展开更多
This study examines the language learning needs and influencing factors of international MBBS students in China, to promote deep learning. Despite compulsory Chinese requirements (Level 4 HSK), a non-immersive environ...This study examines the language learning needs and influencing factors of international MBBS students in China, to promote deep learning. Despite compulsory Chinese requirements (Level 4 HSK), a non-immersive environment often leads to motivation issues. Findings reveal that students perceive their needs as phased, homogeneous yet diverse, and not universal. Daily and professional communication demands, positive teacher-student relationships, and successful language application drive deep learning. Key strategies are proposed, including enhancing teacher quality, developing structured materials that bridge general and medical Chinese, implementing scenario-based teaching that prioritizes speaking/listening, and optimizing the curriculum for continuous exposure and a balanced workload. The ultimate goal is to cultivate competent communicators for medical practice and daily life in China.展开更多
Online interactive learning plays a crucial role in improving online education quality.This grounded theory study examines:(1)what key factors shape EFL learners’online interactive learning,(2)how these factors form ...Online interactive learning plays a crucial role in improving online education quality.This grounded theory study examines:(1)what key factors shape EFL learners’online interactive learning,(2)how these factors form an empirically validated model,and(3)how they interact within this model,through systematic analysis of 9,207 discussion forum posts from a Chinese University MOOC platform.Results demonstrate that learning drive,course structure,teaching competence,interaction behavior,expected outcomes,and online learning context significantly influence EFL online interactive learning.The analysis reveals two key mechanisms:expected outcomes mediate the effects of learning drive(β=0.45),course structure,teaching competence,and interaction behavior(β=0.35)on learning outcomes,while online learning context moderates these relationships(β=0.25).Specifically,learning drive provides intrinsic/extrinsic motivation,whereas course structure,teaching competence,interaction behavior,and expected outcomes collectively enhance interaction quality and sustainability.These findings,derived through rigorous grounded theory methodology involving open,axial,and selective coding of large-scale interaction data,yield three key contributions:(1)a comprehensive theoretical model of EFL online learning dynamics,(2)empirical validation of mediation/moderation mechanisms,and(3)practical strategies for designing scaffolded interaction protocols and adaptive feedback systems.The study establishes that its theoretically saturated model(achieved after analyzing 7,366 posts with 1,841 verification cases)offers educators evidence-based approaches to optimize collaborative interaction in digital EFL environments.展开更多
Biometric characteristics are playing a vital role in security for the last few years.Human gait classification in video sequences is an important biometrics attribute and is used for security purposes.A new framework...Biometric characteristics are playing a vital role in security for the last few years.Human gait classification in video sequences is an important biometrics attribute and is used for security purposes.A new framework for human gait classification in video sequences using deep learning(DL)fusion assisted and posterior probability-based moth flames optimization(MFO)is proposed.In the first step,the video frames are resized and finetuned by two pre-trained lightweight DL models,EfficientNetB0 and MobileNetV2.Both models are selected based on the top-5 accuracy and less number of parameters.Later,both models are trained through deep transfer learning and extracted deep features fused using a voting scheme.In the last step,the authors develop a posterior probabilitybased MFO feature selection algorithm to select the best features.The selected features are classified using several supervised learning methods.The CASIA-B publicly available dataset has been employed for the experimental process.On this dataset,the authors selected six angles such as 0°,18°,90°,108°,162°,and 180°and obtained an average accuracy of 96.9%,95.7%,86.8%,90.0%,95.1%,and 99.7%.Results demonstrate comparable improvement in accuracy and significantly minimize the computational time with recent state-of-the-art techniques.展开更多
We propose a novel type of nonlinear solver acceleration for systems of nonlinear partial differential equations(PDEs)that is based on online/adaptive learning.It is applied in the context of multiphase flow in porous...We propose a novel type of nonlinear solver acceleration for systems of nonlinear partial differential equations(PDEs)that is based on online/adaptive learning.It is applied in the context of multiphase flow in porous media.The proposed method rely on four pillars:(i)dimensionless numbers as input parameters for the machine learning model,(ii)simplified numerical model(two-dimensional)for the offline training,(iii)dynamic control of a nonlinear solver tuning parameter(numerical relaxation),(iv)and online learning for time real-improvement of the machine learning model.This strategy decreases the number of nonlinear iterations by dynamically modifying a single global parameter,the relaxation factor,and by adaptively learning the attributes of each numerical model on-the-run.Furthermore,this work performs a sensitivity study in the dimensionless parameters(machine learning features),assess the efficacy of various machine learning models,demonstrate a decrease in nonlinear iterations using our method in more intricate,realistic three-dimensional models,and fully couple a machine learning model into an open-source multiphase flow simulator achieving up to 85%reduction in computational time.展开更多
After living in China for 33 years,Associate Professor Gu Qingyang of the Lee Kuan Yew School of Public Policy(LKYSPP)at the National University of Singapore(NUS)arrived in Singapore in 1994.Over the past 31 years,he ...After living in China for 33 years,Associate Professor Gu Qingyang of the Lee Kuan Yew School of Public Policy(LKYSPP)at the National University of Singapore(NUS)arrived in Singapore in 1994.Over the past 31 years,he has remained dedicated to building bridges—initially by systematically introducing Singapore’s development experience to China,and later by fostering mutual learning between the two countries.展开更多
Polymers often exhibit multi-state conformational transitions with multiple pathways as temperature varies.However,characterizing the inherent features of these pathways is hindered by the lack of physical characteriz...Polymers often exhibit multi-state conformational transitions with multiple pathways as temperature varies.However,characterizing the inherent features of these pathways is hindered by the lack of physical characterizations that can distinguish various transition pathways between complex and disordered states.In this work,we introduced a machine-learning framework based on spatiotemporal point-cloud neural networks to identify and analyze conformational transition pathways in polymer chains.As a case study,we applied this framework to the temperature-induced unfolding of a single semi-flexible polymer chain,simulated via coarse-grained molecular dynamics.We first combined spatiotemporal point cloud neural networks and contrastive learning to extract features of conformational evolution,and then we employed unsupervised learning methods to cluster distinct transition pathways and unfolding trajectories.Our results reveal that,with increasing temperature,semi-flexible polymer chains exhibit five distinct unfolding pathways:rigid rod→random coil;small toroid→large toroid→hairpin→random coil;rod bundle→hairpin→random coil;hairpin→random coil;and tailed structure→random coil.We further calculated the structural order parameters of those typical conformations with distinct transition pathways,we distincted five transition mechanisms,including the straightening of rigid rods,tightening of small rings,expansion of hairpin ends,symmetrization of rod bundles,and retraction of tailed structures.These findings demonstrate that our framework presents a promising data-driven approach for analyzing complex conformational transitions in disordered polymers,which might be potentially extendable to other heterogeneous systems like intrinsically disordered proteins.展开更多
基金supported by National Natural Science Foundation of China(62466045)Inner Mongolia Natural Science Foundation Project(2021LHMS06003)Inner Mongolia University Basic Research Business Fee Project(114).
文摘Graph Federated Learning(GFL)has shown great potential in privacy protection and distributed intelligence through distributed collaborative training of graph-structured data without sharing raw information.However,existing GFL approaches often lack the capability for comprehensive feature extraction and adaptive optimization,particularly in non-independent and identically distributed(NON-IID)scenarios where balancing global structural understanding and local node-level detail remains a challenge.To this end,this paper proposes a novel framework called GFL-SAR(Graph Federated Collaborative Learning Framework Based on Structural Amplification and Attention Refinement),which enhances the representation learning capability of graph data through a dual-branch collaborative design.Specifically,we propose the Structural Insight Amplifier(SIA),which utilizes an improved Graph Convolutional Network(GCN)to strengthen structural awareness and improve modeling of topological patterns.In parallel,we propose the Attentive Relational Refiner(ARR),which employs an enhanced Graph Attention Network(GAT)to perform fine-grained modeling of node relationships and neighborhood features,thereby improving the expressiveness of local interactions and preserving critical contextual information.GFL-SAR effectively integrates multi-scale features from every branch via feature fusion and federated optimization,thereby addressing existing GFL limitations in structural modeling and feature representation.Experiments on standard benchmark datasets including Cora,Citeseer,Polblogs,and Cora_ML demonstrate that GFL-SAR achieves superior performance in classification accuracy,convergence speed,and robustness compared to existing methods,confirming its effectiveness and generalizability in GFL tasks.
基金financial support provided by the Natural Science Foundation of Hebei Province,China(No.E2024105036)the Tangshan Talent Funding Project,China(Nos.B202302007 and A2021110015)+1 种基金the National Natural Science Foundation of China(No.52264042)the Australian Research Council(No.IH230100010)。
文摘Automated classification of gas flow states in blast furnaces using top-camera imagery typically demands a large volume of labeled data,whose manual annotation is both labor-intensive and cost-prohibitive.To mitigate this challenge,we present an enhanced semi-supervised learning approach based on the Mean Teacher framework,incorporating a novel feature loss module to maximize classification performance with limited labeled samples.The model studies show that the proposed model surpasses both the baseline Mean Teacher model and fully supervised method in accuracy.Specifically,for datasets with 20%,30%,and 40%label ratios,using a single training iteration,the model yields accuracies of 78.61%,82.21%,and 85.2%,respectively,while multiple-cycle training iterations achieves 82.09%,81.97%,and 81.59%,respectively.Furthermore,scenario-specific training schemes are introduced to support diverse deployment need.These findings highlight the potential of the proposed technique in minimizing labeling requirements and advancing intelligent blast furnace diagnostics.
基金National Key Research and Development Program of China,No.2023YFC3006704National Natural Science Foundation of China,No.42171047CAS-CSIRO Partnership Joint Project of 2024,No.177GJHZ2023097MI。
文摘Accurate prediction of flood events is important for flood control and risk management.Machine learning techniques contributed greatly to advances in flood predictions,and existing studies mainly focused on predicting flood resource variables using single or hybrid machine learning techniques.However,class-based flood predictions have rarely been investigated,which can aid in quickly diagnosing comprehensive flood characteristics and proposing targeted management strategies.This study proposed a prediction approach of flood regime metrics and event classes coupling machine learning algorithms with clustering-deduced membership degrees.Five algorithms were adopted for this exploration.Results showed that the class membership degrees accurately determined event classes with class hit rates up to 100%,compared with the four classes clustered from nine regime metrics.The nonlinear algorithms(Multiple Linear Regression,Random Forest,and least squares-Support Vector Machine)outperformed the linear techniques(Multiple Linear Regression and Stepwise Regression)in predicting flood regime metrics.The proposed approach well predicted flood event classes with average class hit rates of 66.0%-85.4%and 47.2%-76.0%in calibration and validation periods,respectively,particularly for the slow and late flood events.The predictive capability of the proposed prediction approach for flood regime metrics and classes was considerably stronger than that of hydrological modeling approach.
基金supported by Key Science and Technology Program of Henan Province,China(Grant Nos.242102210147,242102210027)Fujian Province Young and Middle aged Teacher Education Research Project(Science and Technology Category)(No.JZ240101)(Corresponding author:Dong Yuan).
文摘Vehicle Edge Computing(VEC)and Cloud Computing(CC)significantly enhance the processing efficiency of delay-sensitive and computation-intensive applications by offloading compute-intensive tasks from resource-constrained onboard devices to nearby Roadside Unit(RSU),thereby achieving lower delay and energy consumption.However,due to the limited storage capacity and energy budget of RSUs,it is challenging to meet the demands of the highly dynamic Internet of Vehicles(IoV)environment.Therefore,determining reasonable service caching and computation offloading strategies is crucial.To address this,this paper proposes a joint service caching scheme for cloud-edge collaborative IoV computation offloading.By modeling the dynamic optimization problem using Markov Decision Processes(MDP),the scheme jointly optimizes task delay,energy consumption,load balancing,and privacy entropy to achieve better quality of service.Additionally,a dynamic adaptive multi-objective deep reinforcement learning algorithm is proposed.Each Double Deep Q-Network(DDQN)agent obtains rewards for different objectives based on distinct reward functions and dynamically updates the objective weights by learning the value changes between objectives using Radial Basis Function Networks(RBFN),thereby efficiently approximating the Pareto-optimal decisions for multiple objectives.Extensive experiments demonstrate that the proposed algorithm can better coordinate the three-tier computing resources of cloud,edge,and vehicles.Compared to existing algorithms,the proposed method reduces task delay and energy consumption by 10.64%and 5.1%,respectively.
基金supported by Systematic Major Project of Shuohuang Railway Development Co.,Ltd.,National Energy Group(Grant Number:SHTL-23-31)Beijing Natural Science Foundation(U22B2027).
文摘In the realm of Intelligent Railway Transportation Systems,effective multi-party collaboration is crucial due to concerns over privacy and data silos.Vertical Federated Learning(VFL)has emerged as a promising approach to facilitate such collaboration,allowing diverse entities to collectively enhance machine learning models without the need to share sensitive training data.However,existing works have highlighted VFL’s susceptibility to privacy inference attacks,where an honest but curious server could potentially reconstruct a client’s raw data from embeddings uploaded by the client.This vulnerability poses a significant threat to VFL-based intelligent railway transportation systems.In this paper,we introduce SensFL,a novel privacy-enhancing method to against privacy inference attacks in VFL.Specifically,SensFL integrates regularization of the sensitivity of embeddings to the original data into the model training process,effectively limiting the information contained in shared embeddings.By reducing the sensitivity of embeddings to the original data,SensFL can effectively resist reverse privacy attacks and prevent the reconstruction of the original data from the embeddings.Extensive experiments were conducted on four distinct datasets and three different models to demonstrate the efficacy of SensFL.Experiment results show that SensFL can effectively mitigate privacy inference attacks while maintaining the accuracy of the primary learning task.These results underscore SensFL’s potential to advance privacy protection technologies within VFL-based intelligent railway systems,addressing critical security concerns in collaborative learning environments.
基金supported by the National Natural Science Foundation of China(No.U21A20290)Guangdong Basic and Applied Basic Research Foundation(No.2022A1515011656)+2 种基金the Projects of Talents Recruitment of GDUPT(No.2023rcyj1003)the 2022“Sail Plan”Project of Maoming Green Chemical Industry Research Institute(No.MMGCIRI2022YFJH-Y-024)Maoming Science and Technology Project(No.2023382).
文摘The presence of aluminum(Al^(3+))and fluoride(F^(−))ions in the environment can be harmful to ecosystems and human health,highlighting the need for accurate and efficient monitoring.In this paper,an innovative approach is presented that leverages the power of machine learning to enhance the accuracy and efficiency of fluorescence-based detection for sequential quantitative analysis of aluminum(Al^(3+))and fluoride(F^(−))ions in aqueous solutions.The proposed method involves the synthesis of sulfur-functionalized carbon dots(C-dots)as fluorescence probes,with fluorescence enhancement upon interaction with Al^(3+)ions,achieving a detection limit of 4.2 nmol/L.Subsequently,in the presence of F^(−)ions,fluorescence is quenched,with a detection limit of 47.6 nmol/L.The fingerprints of fluorescence images are extracted using a cross-platform computer vision library in Python,followed by data preprocessing.Subsequently,the fingerprint data is subjected to cluster analysis using the K-means model from machine learning,and the average Silhouette Coefficient indicates excellent model performance.Finally,a regression analysis based on the principal component analysis method is employed to achieve more precise quantitative analysis of aluminum and fluoride ions.The results demonstrate that the developed model excels in terms of accuracy and sensitivity.This groundbreaking model not only showcases exceptional performance but also addresses the urgent need for effective environmental monitoring and risk assessment,making it a valuable tool for safeguarding our ecosystems and public health.
基金supported by the National Key Research and Development Plan of the Ministry of Science and Technology,China(Grant No.:2022YFE0125300)the National Natural Science Foundation of China(Grant No:81690262)+2 种基金the National Science and Technology Major Project,China(Grant No.:2017ZX09201004-021)the Open Project of National facility for Translational Medicine(Shanghai),China(Grant No.:TMSK-2021-104)Shanghai Jiao Tong University STAR Grant,China(Grant Nos.:YG2022ZD024 and YG2022QN111).
文摘Liposomes serve as critical carriers for drugs and vaccines,with their biological effects influenced by their size.The microfluidic method,renowned for its precise control,reproducibility,and scalability,has been widely employed for liposome preparation.Although some studies have explored factors affecting liposomal size in microfluidic processes,most focus on small-sized liposomes,predominantly through experimental data analysis.However,the production of larger liposomes,which are equally significant,remains underexplored.In this work,we thoroughly investigate multiple variables influencing liposome size during microfluidic preparation and develop a machine learning(ML)model capable of accurately predicting liposomal size.Experimental validation was conducted using a staggered herringbone micromixer(SHM)chip.Our findings reveal that most investigated variables significantly influence liposomal size,often interrelating in complex ways.We evaluated the predictive performance of several widely-used ML algorithms,including ensemble methods,through cross-validation(CV)for both lipo-some size and polydispersity index(PDI).A standalone dataset was experimentally validated to assess the accuracy of the ML predictions,with results indicating that ensemble algorithms provided the most reliable predictions.Specifically,gradient boosting was selected for size prediction,while random forest was employed for PDI prediction.We successfully produced uniform large(600 nm)and small(100 nm)liposomes using the optimised experimental conditions derived from the ML models.In conclusion,this study presents a robust methodology that enables precise control over liposome size distribution,of-fering valuable insights for medicinal research applications.
基金supported by the National Natural Science Foundation of China under Grant No.U20A20182.
文摘As an effective strategy to address urban traffic congestion,traffic flow prediction has gained attention from Federated-Learning(FL)researchers due FL’s ability to preserving data privacy.However,existing methods face challenges:some are too simplistic to capture complex traffic patterns effectively,and others are overly complex,leading to excessive communication overhead between cloud and edge devices.Moreover,the problem of single point failure limits their robustness and reliability in real-world applications.To tackle these challenges,this paper proposes a new method,CMBA-FL,a Communication-Mitigated and Blockchain-Assisted Federated Learning model.First,CMBA-FL improves the client model’s ability to capture temporal traffic patterns by employing the Encoder-Decoder framework for each edge device.Second,to reduce the communication overhead during federated learning,we introduce a verification method based on parameter update consistency,avoiding unnecessary parameter updates.Third,to mitigate the risk of a single point of failure,we integrate consensus mechanisms from blockchain technology.To validate the effectiveness of CMBA-FL,we assess its performance on two widely used traffic datasets.Our experimental results show that CMBA-FL reduces prediction error by 11.46%,significantly lowers communication overhead,and improves security.
基金supported by the Sichuan Province Nursing Scientific Research Project Plan(H23022)the 2022 Municipal-University Science and Technology Strategic Cooperation Special Fund of Nanchong Science and Technology Bureau(22SXQT0222)。
文摘Objective Accurately identifying the key influencing factors of psychological birth trauma in primiparous women is crucial for implementing effective preventive and intervention measures.This study aimed to develop and validate an interpretable machine learning prediction model for identifying the key influencing factors of psychological birth trauma in primiparous women.Methods A multicenter cross-sectional study was conducted on primiparous women in four tertiary hospitals in Sichuan Province,southwestern China,from December 2023 to March 2024.The Childbirth Trauma Index was used in assessing psychological birth trauma in primiparous women.Data were collected and randomly divided into a training set(80%,n=289)and a testing set(20%,n=73).Six different machine learning models were trained and tested.Training and prediction were conducted using six machine learning models included Linear Regression,Support Vector Regression,Multilayer Perceptron Regression,eXtreme Gradient Boosting Regression,Random Forest Regression,and Adaptive Boosting Regression.The optimal model was selected based on various performance metrics,and its predictive results were interpreted using SHapley Additive exPlanations(SHAP)and accumulated local effects(ALE).Results Among the six machine learning models,the Multilayer Perceptron Regression model exhibited the best overall performance in the testing set(MAE=3.977,MSE=24.832,R2=0.507,EVS=0.524,RMSE=4.983).In the testing set,the R2 and EVS of the Multilayer Perceptron Regression model increased by 8.3%and 1.2%,respectively,compared to the traditional linear regression model.Meanwhile,the MAE,MSE,and RMSE decreased by 0.4%,7.3%,and 3.7%,respectively,compared to the traditional linear regression model.The SHAP analysis indicated that intrapartum pain,anxiety,postpartum pain,resilience,and planned pregnancy are the most critical influencing factors of psychological birth trauma in primiparous women.The ALE analysis indicated that higher intrapartum pain,anxiety,and postpartum pain scores are risk factors,while higher resilience scores are protective factors.Conclusions Interpretable machine learning prediction models can identify the key influencing factors of psychological birth trauma in primiparous women.SHAP and ALE analyses based on the Multilayer Perceptron Regression model can help healthcare providers understand the complex decision-making logic within a prediction model.This study provides a scientific basis for the early prevention and personalized intervention of psychological birth trauma in primiparous women.
基金Funded by the Natural Science Foundation of China(No.52109168)。
文摘In order to study the characteristics of pure fly ash-based geopolymer concrete(PFGC)conveniently,we used a machine learning method that can quantify the perception of characteristics to predict its compressive strength.In this study,505 groups of data were collected,and a new database of compressive strength of PFGC was constructed.In order to establish an accurate prediction model of compressive strength,five different types of machine learning networks were used for comparative analysis.The five machine learning models all showed good compressive strength prediction performance on PFGC.Among them,R2,MSE,RMSE and MAE of decision tree model(DT)are 0.99,1.58,1.25,and 0.25,respectively.While R2,MSE,RMSE and MAE of random forest model(RF)are 0.97,5.17,2.27 and 1.38,respectively.The two models have high prediction accuracy and outstanding generalization ability.In order to enhance the interpretability of model decision-making,we used importance ranking to obtain the perception of machine learning model to 13 variables.These 13 variables include chemical composition of fly ash(SiO_(2)/Al_(2)O_(3),Si/Al),the ratio of alkaline liquid to the binder,curing temperature,curing durations inside oven,fly ash dosage,fine aggregate dosage,coarse aggregate dosage,extra water dosage and sodium hydroxide dosage.Curing temperature,specimen ages and curing durations inside oven have the greatest influence on the prediction results,indicating that curing conditions have more prominent influence on the compressive strength of PFGC than ordinary Portland cement concrete.The importance of curing conditions of PFGC even exceeds that of the concrete mix proportion,due to the low reactivity of pure fly ash.
文摘Teacher–student relationships play a vital role in improving college students’academic performance and the quality of higher education.However,empirical studies with substantial data-driven insights remain limited.To address this gap,this study collected 3278 questionnaires from seven universities across four provinces in China to analyze the key factors affecting college students’academic performance.A machine learning framework,CQFOA-KELM,was developed by enhancing the Fruit Fly Optimization Algorithm(FOA)with Covariance Matrix Adaptation Evolution Strategy(CMAES)and Quadratic Approximation(QA).CQFOA significantly improved population diversity and was validated on the IEEE CEC2017 benchmark functions.The CQFOA-KELM model achieved an accuracy of 98.15%and a sensitivity of 98.53%in predicting college students’academic performance.Additionally,it effectively identified the key factors influencing academic performance through the feature selection process.
基金supported in part by the National Key Research and Development Program of China under Grant No.2021YFF0901300in part by the National Natural Science Foundation of China under Grant Nos.62173076 and 72271048.
文摘The distributed permutation flow shop scheduling problem(DPFSP)has received increasing attention in recent years.The iterated greedy algorithm(IGA)serves as a powerful optimizer for addressing such a problem because of its straightforward,single-solution evolution framework.However,a potential draw-back of IGA is the lack of utilization of historical information,which could lead to an imbalance between exploration and exploitation,especially in large-scale DPFSPs.As a consequence,this paper develops an IGA with memory and learning mechanisms(MLIGA)to efficiently solve the DPFSP targeted at the mini-malmakespan.InMLIGA,we incorporate a memory mechanism to make a more informed selection of the initial solution at each stage of the search,by extending,reconstructing,and reinforcing the information from previous solutions.In addition,we design a twolayer cooperative reinforcement learning approach to intelligently determine the key parameters of IGA and the operations of the memory mechanism.Meanwhile,to ensure that the experience generated by each perturbation operator is fully learned and to reduce the prior parameters of MLIGA,a probability curve-based acceptance criterion is proposed by combining a cube root function with custom rules.At last,a discrete adaptive learning rate is employed to enhance the stability of the memory and learningmechanisms.Complete ablation experiments are utilized to verify the effectiveness of the memory mechanism,and the results show that this mechanism is capable of improving the performance of IGA to a large extent.Furthermore,through comparative experiments involving MLIGA and five state-of-the-art algorithms on 720 benchmarks,we have discovered that MLI-GA demonstrates significant potential for solving large-scale DPFSPs.This indicates that MLIGA is well-suited for real-world distributed flow shop scheduling.
基金supported by grants from the Zhejiang Provincial Natural Science Foundation(LGJ22H180001)Zhejiang Medical and Health Science and Technology Project(2021KY249)the National Key R&D Program of China(2017YFC1310000).
文摘Attempts have been made to modulate motor sequence learning(MSL)through repetitive transcranial magnetic stimulation,targeting different sites within the sensorimotor network.However,the target with the optimum modulatory effect on neural plasticity associated with MSL remains unclarified.This study was therefore designed to compare the role of the left primary motor cortex and the left supplementary motor area proper(SMAp)in modulating MSL across different complexity levels and for both hands,as well as the associated neuroplasticity by applying intermittent theta burst stimulation together with the electroencephalogram and concurrent transcranial magnetic stimulation.Our data demonstrated the role of SMAp stimulation in modulating neural communication to support MSL,which is achieved by facilitating regional activation and orchestrating neural coupling across distributed brain regions,particularly in interhemispheric connections.These findings may have important clinical implications,particularly for motor rehabilitation in populations such as post-stroke patients.
基金supported by the 2024“Special Research Project on the Application of Artificial Intelligence in Empowering Teaching and Education”of Zhejiang Province Association of Higher Education(KT2024165).
文摘While artificial intelligence(AI)shows promise in education,its real-world effectiveness in specific settings like blended English as a Foreign Language(EFL)learning needs closer examination.This study investigated the impact of a blended teaching model incorporating AI tools on the Superstar Learning Platform for Chinese university EFL students.Using a mixed-methods approach,60 first-year students were randomized into an experimental group(using the AI-enhanced model)and a control group(traditional instruction)for 16 weeks.Data included test scores,learning behaviors(duration,task completion),satisfaction surveys,and interviews.Results showed the experimental group significantly outperformed the control group on post-tests and achieved larger learning gains.These students also demonstrated greater engagement through longer study times and higher task completion rates,and reported significantly higher satisfaction.Interviews confirmed these findings,with students attributing benefits to the model’s personalized guidance,structured content presentation(knowledge graphs),immediate responses,flexibility,and varied interaction methods.However,limitations were noted,including areas where the platform’s AI could be improved(e.g.,for assessing speaking/translation)and ongoing challenges with student self-discipline.The study concludes that this AI-enhanced blended model significantly improved student performance,engagement,and satisfaction in this EFL context.The findings offer practical insights for educators and platform developers,suggesting AI integration holds significant potential while highlighting areas for refinement.
基金supported by the Chinese Testing International Co.,Ltd.(CTI)under Grant CTI2022B01.
文摘This study examines the language learning needs and influencing factors of international MBBS students in China, to promote deep learning. Despite compulsory Chinese requirements (Level 4 HSK), a non-immersive environment often leads to motivation issues. Findings reveal that students perceive their needs as phased, homogeneous yet diverse, and not universal. Daily and professional communication demands, positive teacher-student relationships, and successful language application drive deep learning. Key strategies are proposed, including enhancing teacher quality, developing structured materials that bridge general and medical Chinese, implementing scenario-based teaching that prioritizes speaking/listening, and optimizing the curriculum for continuous exposure and a balanced workload. The ultimate goal is to cultivate competent communicators for medical practice and daily life in China.
文摘Online interactive learning plays a crucial role in improving online education quality.This grounded theory study examines:(1)what key factors shape EFL learners’online interactive learning,(2)how these factors form an empirically validated model,and(3)how they interact within this model,through systematic analysis of 9,207 discussion forum posts from a Chinese University MOOC platform.Results demonstrate that learning drive,course structure,teaching competence,interaction behavior,expected outcomes,and online learning context significantly influence EFL online interactive learning.The analysis reveals two key mechanisms:expected outcomes mediate the effects of learning drive(β=0.45),course structure,teaching competence,and interaction behavior(β=0.35)on learning outcomes,while online learning context moderates these relationships(β=0.25).Specifically,learning drive provides intrinsic/extrinsic motivation,whereas course structure,teaching competence,interaction behavior,and expected outcomes collectively enhance interaction quality and sustainability.These findings,derived through rigorous grounded theory methodology involving open,axial,and selective coding of large-scale interaction data,yield three key contributions:(1)a comprehensive theoretical model of EFL online learning dynamics,(2)empirical validation of mediation/moderation mechanisms,and(3)practical strategies for designing scaffolded interaction protocols and adaptive feedback systems.The study establishes that its theoretically saturated model(achieved after analyzing 7,366 posts with 1,841 verification cases)offers educators evidence-based approaches to optimize collaborative interaction in digital EFL environments.
基金King Saud University,Grant/Award Number:RSP2024R157。
文摘Biometric characteristics are playing a vital role in security for the last few years.Human gait classification in video sequences is an important biometrics attribute and is used for security purposes.A new framework for human gait classification in video sequences using deep learning(DL)fusion assisted and posterior probability-based moth flames optimization(MFO)is proposed.In the first step,the video frames are resized and finetuned by two pre-trained lightweight DL models,EfficientNetB0 and MobileNetV2.Both models are selected based on the top-5 accuracy and less number of parameters.Later,both models are trained through deep transfer learning and extracted deep features fused using a voting scheme.In the last step,the authors develop a posterior probabilitybased MFO feature selection algorithm to select the best features.The selected features are classified using several supervised learning methods.The CASIA-B publicly available dataset has been employed for the experimental process.On this dataset,the authors selected six angles such as 0°,18°,90°,108°,162°,and 180°and obtained an average accuracy of 96.9%,95.7%,86.8%,90.0%,95.1%,and 99.7%.Results demonstrate comparable improvement in accuracy and significantly minimize the computational time with recent state-of-the-art techniques.
基金MUFFINS,MUltiphase Flow-induced Fluid-flexible structure InteractioN in Subsea applications(EP/P033180/1)the PREMIERE programme grant(EP/T000414/1)SMARTRES,Smart assessment,management and optimization of urban geothermal resources(NE/X005607/1).
文摘We propose a novel type of nonlinear solver acceleration for systems of nonlinear partial differential equations(PDEs)that is based on online/adaptive learning.It is applied in the context of multiphase flow in porous media.The proposed method rely on four pillars:(i)dimensionless numbers as input parameters for the machine learning model,(ii)simplified numerical model(two-dimensional)for the offline training,(iii)dynamic control of a nonlinear solver tuning parameter(numerical relaxation),(iv)and online learning for time real-improvement of the machine learning model.This strategy decreases the number of nonlinear iterations by dynamically modifying a single global parameter,the relaxation factor,and by adaptively learning the attributes of each numerical model on-the-run.Furthermore,this work performs a sensitivity study in the dimensionless parameters(machine learning features),assess the efficacy of various machine learning models,demonstrate a decrease in nonlinear iterations using our method in more intricate,realistic three-dimensional models,and fully couple a machine learning model into an open-source multiphase flow simulator achieving up to 85%reduction in computational time.
文摘After living in China for 33 years,Associate Professor Gu Qingyang of the Lee Kuan Yew School of Public Policy(LKYSPP)at the National University of Singapore(NUS)arrived in Singapore in 1994.Over the past 31 years,he has remained dedicated to building bridges—initially by systematically introducing Singapore’s development experience to China,and later by fostering mutual learning between the two countries.
基金financially supported by the National Key R&D Program of China(No.2022YFB3707303)the National Natural Science Foundation of China(No.52293471)。
文摘Polymers often exhibit multi-state conformational transitions with multiple pathways as temperature varies.However,characterizing the inherent features of these pathways is hindered by the lack of physical characterizations that can distinguish various transition pathways between complex and disordered states.In this work,we introduced a machine-learning framework based on spatiotemporal point-cloud neural networks to identify and analyze conformational transition pathways in polymer chains.As a case study,we applied this framework to the temperature-induced unfolding of a single semi-flexible polymer chain,simulated via coarse-grained molecular dynamics.We first combined spatiotemporal point cloud neural networks and contrastive learning to extract features of conformational evolution,and then we employed unsupervised learning methods to cluster distinct transition pathways and unfolding trajectories.Our results reveal that,with increasing temperature,semi-flexible polymer chains exhibit five distinct unfolding pathways:rigid rod→random coil;small toroid→large toroid→hairpin→random coil;rod bundle→hairpin→random coil;hairpin→random coil;and tailed structure→random coil.We further calculated the structural order parameters of those typical conformations with distinct transition pathways,we distincted five transition mechanisms,including the straightening of rigid rods,tightening of small rings,expansion of hairpin ends,symmetrization of rod bundles,and retraction of tailed structures.These findings demonstrate that our framework presents a promising data-driven approach for analyzing complex conformational transitions in disordered polymers,which might be potentially extendable to other heterogeneous systems like intrinsically disordered proteins.
